he aim of the training program that we are proposing is to educate a generation of scientists who have the knowledge required to identify important biological problems and the expertise required to develop and apply advanced computational methods towards their solution. The scientist we envision will be an integral part of a discipline involving computational science but will be comfortable attending a meeting of experimental biologists as well. Indeed we believe that it is essential that the next generation of computational biologists have a deep understanding of biology. In addition it is important that they have familiarity with more than one computational discipline. For example, it would be desirable if computer scientists working on new algorithms to classify proteins would understand structural biology and have expertise in computational studies of protein structure and function as well. Integrating different computational disciplines is a challenging goal in its own right and is made more complex by the fact that the unifying theme is biological. We will ensure that our trainees have a deep understanding of experimental biology, have expertise in at least one area of computational biology and have familiarity with other areas.